X hits on this document

PDF document

The split-fovea model of visual word - page 6 / 78

207 views

0 shares

0 downloads

0 comments

6 / 78

Chapter 1

Introduction and Background

The first section of this chapter will provide a brief introduction and overview of the current paper, while the second section will give the background needed for the current work.

1.1

Introduction

Connectionist networks have been used extensively in the domain of visual word recognition and reading. There have been a number of connectionist networks in re- cent years which model the reading behaviour of people. An interesting model is the split-fovea model (Shillcock and Monaghan, 2003). This model integrates the facts that the brain and the fovea are both split into its architecture by having two separate inputs for the right and left visual fields and two separate hidden layers for the two hemispheres. Doing this allows one to present words to the network at different fixa- tion positions. Because of this possibility the network is well suited for experimenting with words being presented at different fixation positions. The current work involves comparing two networks. The first one is a control network similar to the network used by Shillcock and Monaghan (2003). The second network (the fixation net) on the other

1

Document info
Document views207
Page views215
Page last viewedWed Dec 07 06:05:50 UTC 2016
Pages78
Paragraphs3087
Words17414

Comments